The Economic Case for Restricted Immigration
Mark Krikorian, Center for Immigration Studies
Dowload PDF Version
We are considering whether there is an economic case for open borders. The question before us is not the utility of any particular visa program or bill before Congress. Rather, we are inquiring as to the economic consequences of lifting all controls on the settlement of foreigners in the United States.
Immigration is different from other aspects of policy precisely because it involves the arrival of outsiders – non-members – into our community. While all men, everywhere, are indeed created equal and endowed by their Creator with certain unalienable rights, our predecessors assumed among the powers of the earth a separate and equal station for the American people as a distinct political community to secure those rights.
The Constitution does not begin, "We the People of the Human Race". Rather it is "We the People of the United States" who have ordained and established this particular political order to secure the blessings of liberty for ourselves and our posterity.
Open borders, therefore, as an ideological principle, as a moral imperative -- as it is often presented by some of the more Jacobin libertarians – is contrary to the very idea of self-government. If we the people do not have authority to decide which outsiders may become members of our national community, and on what terms, then in what sense can that national community be described as sovereign?
That having been said, unlimited immigration might nonetheless be a choice that the elected representatives of the sovereign American people would want to make. And so let's look at the economic consequences of immigration.
The first is the effect immigration has on the aggregate size of the U.S. economy. If we think of the economy as being comprised of capital and labor, then adding more labor obviously makes the overall GDP larger. Immigrants account for about 16 percent of all workers. Government data show that their combined earnings in 2010 were over $900 billion dollars; this figure understates their labor incomes, so it seems certain that the GDP of the United States is at minimum one trillion dollars larger because of the presence of immigrants.
But a larger economy by itself is not necessarily a benefit to the native-born population, and that's the only yardstick for measuring the desirability of any government policy. The literature on the economics of immigration makes clear that almost all of the increased economic activity that immigrants generate goes to the immigrants themselves in the form of wages and benefits – as it should, since they're the ones doing the work.
To determine the benefit of immigration on the existing population, the only relevant measure is per-capita GDP, particularly the per-capita GDP of the pre-existing or native-born population. We can see the importance of per-capita GDP versus aggregate GDP by simply noting that Canada and Indonesia have similar-sized economies, and yet no one would say that the average Indonesian is as rich as the average Canadian, since there are seven times more Indonesians than Canadians.
This, then, is a second, more important, consideration in the economics of immigration: how much do native-born Americans benefit from it? There's a standard way of calculating the benefit from immigration – also referred to the as the "immigrant surplus" that goes to the existing population.
A 1997 study by the National Academy of Sciences (NAS), authored by many of the top economists in the field, summarizes the formula for calculating the benefit from immigration. The NAS study updates an earlier study by the nation’s top immigration economist, George Borjas of Harvard. And in 2007 the President Bush's Council of Economic Advisers (CEA) again used the same formula to estimate the benefit of immigration to Americans. A blog by professor Borjas has a clear non-technical explanation of the calculation.
What they found was that the net gain from immigration equals about one-quarter of one percent of the economy. If GDP is $15 trillion, then the net benefit would be $40 billion.
Three important points:
- First, the net effect of immigration on the existing population is positive overall, though not for all workers.
- Second, at only about one-quarter of one percent of GDP, the benefit is trivial relative to the size of the economy.
- Third, and this is the most important point for policymakers, the net benefit of immigration comes from lowering labor costs – lowering the wages of American workers in competition with immigrants. In other words, the bigger the reduction in wages for certain American workers, the bigger the net benefit for the nation overall.
The same model can be used to estimate the wage losses suffered American workers in the same occupations as immigrants; they amount to 2.8 percent of GDP, or $420 billion in wage losses in a $15 trillion economy suffered by American workers because of immigration. There is nothing particularly controversial about this estimate and it stems from the same basic economic formula as the one used to calculate the immigrant surplus.
The money that would have gone to workers as wages if there had been no immigration does not vanish into thin air. It maybe go to more skilled workers who are not in competition with immigrants, some may be retained by owners of capital as higher profits, or passed on to consumers in the form of lower prices. Thus the native-bon “winners” from immigration gain $460 billion, the losers lose $420 billion, for a net gain to the native-born of $40 billion.
Thus immigration serves as a government-created vehicle for redistribution of wealth from American blue-collar workers to the owners of capital and American high-skilled workers who do not compete with immigrants.
So policymakers are faced with what is essentially a moral question – is it just to reduce the wages of American-born construction workers, nannies, maids, dishwashers, cooks, and meat packers to increase the wages of lawyers, journalists, and business owners. Guess how it turns out?
And it's important to note that this standard model of calculating the net benefit if immigration takes no account of the “marginal utility of income.” That is to say, it doesn't consider that those who lose from immigration generally have less money to begin with than those who gain from immigration, so the dollar they lose is worth more to them than the dollar gained by the winners. If this were incorporated into the model, then even the small net benefit from immigration might entirely disappear.
The standard economic approach to measuring the benefits of immigration is not the only way to think about the issue. One can make different assumptions and come up with different results. But if you do so then your results will reflect your assumptions and this can cut both ways. For example, two trade economists at Columbia University, Donald Davis and David Weinstein, have developed a model that incorporates the relatively high level of technology in the United States. They find that although immigration increases the size of the U.S. economy and the income of the immigrants, all native-born Americans lose from immigration. That is, there is no net gain from immigration to natives and the wages losses for Americans total nearly $70 billion dollars. On the other hand, economist Giovanni Peri at the University of California, Davis, has created a model that assumes that immigrants don’t compete with immigrants at all – they're “complements” and as a result they slightly increase the wages of natives. But both approaches deviations from the standard model and reflect the underlying assumptions of their different approaches.
One other point that maybe worth making. Whatever you think of the economic impact of immigration, we can say that in general immigration has increased the supply of workers especially in occupations that require relatively less education. To be sure there are some highly skilled occupations that have large concentrations of immigrants as well. Nonetheless, immigration has had its largest impact on non-supervisory occupations in food service, construction, child care, agriculture, building cleaning and maintenance, and light manufacturing. There is widespread agreement that over the last three decades – just as immigration has increased enormously – these occupations, and less-educated workers in general who do this kind of work, have seen a significant decline in their real wages and a very large decline relative to college graduates. Real wages for male high school dropouts have declined 22 percent since the late 1970s and real wages for male high grads are down 10 percent. There are no doubt other factors contributing to this decline, but such a decline is certainly what we would expect if immigration was adversely impacting American workers.
So, immigration increases the overall size of the economy, and it creates a small net benefit for the native-born by driving down the wages of less-educated Americans and distributing that benefit to the rest of society. The third economic factor to consider is the effect of immigration on government spending.
Government now spends a far larger share of national income than during prior waves of immigration – something like 40 percent of GDP – and while I hope this will be reduced somewhat by saner policies in Washington and the state capitals, the basic reality of a large tax-supported government sector is not going away. This includes welfare spending, of course, but also spending on schools, roads, criminal justice, and other tax-supported activities.
The problem this poses was summed up by Milton Friedman: “It’s just obvious that you can’t have free immigration and a welfare state.”
Mass immigration – much less completely open immigration – will inevitably be dominated by the less-educated, the less-skilled, the poor. Importing millions of poor people with large families – as we've done for the past several decades – means that, by definition, they will pay relatively little in taxes but make heavy use of government services. This is true not because of any moral defect in the immigrants or any meaningful differences between today’s immigrants and those of the past. Instead, it is we who have changed – our modern society embraces a larger role for government, expecting it to underwrite a broad system of social provision for the poor, education for the young, support for the elderly, a large portion of the nation’s medical care (even without Obamacare), and many other functions.
And, in fact, even without open borders, our current immigration system, which admits about 1.1 million legal immigrants each year, in addition to several hundred thousand illegal aliens, has imported a poverty class that is supported by taxpayers. In 2010, immigrants and their young children account for one in four persons in poverty in the United States. When you look at what is called "in or near poverty", i.e., 200% of poverty or less, which is the threshold for eligibility for many welfare programs, you see than nearly half of immigrants and their young children are in or near poverty – two-thirds of Mexicans, the largest immigrant group by far.
These high levels of poverty are not due to laziness – immigrants have about the same labor force participation rate as the native-born. Rather it's due to the fact that they have low levels of education – one quarter of all immigrants in the labor force do not have high-school degrees, more than quadruple the rate among the native born – and they've come to a post-industrial, knowledge-based economy that, as I mentioned before, has seen wages for less-skilled workers overall actually decline over the past several decades.
As a result of these low incomes, immigrants pay relatively less in taxes and make relatively greater use of welfare. Among all immigrant households, 36 percent receive at least one means-tested government benefit. Among immigrant households from Latin America, 51 percent use welfare, and for Mexicans, far and away the largest group, 57 percent use welfare.
Some might reply that the solution, then, would be to end welfare and drastically shrink the range of taxpayer-provided services more generally. This is sometimes expressed as "immigration si, welfare no". Well, I'll make a deal with those who feel that way – first, abolish welfare, abolish Medicare, abolish public education, totally privatize Social Security, then give me a call and we'll talk about immigration. But even Nobel Prize-winning economist Gary Becker, who would prefer open borders in theory, has written that "Experience demonstrates that in our political system, it is impossible to prevent immigrants, even those here illegally, to gain access to these benefits. I believe that with unlimited immigration, many would come mainly because they are attracted by these government benefits, and they would then be voting to influence future government spending and other public policies."
And that last observation leads me to the final economic impact of high levels of immigration – it reshapes the political and social environment in ways that undermine free enterprise and economic dynamism.
Leftists understand this quite well and have been quite open about the politically transformative effects of immigration. Eliseo Medina, vice president of the SEIU and an honorary chairman of the Democratic Socialists of America, said this at a left-wing gathering:
The progressive community needs to solidly be on the side of immigrants. They will solidify and expand the progressive coalition for the future... We will create a governing coalition for the long term not just for an election cycle.
A few years earlier, Rep. Barney Frank of Massachusetts made the same point a little more circumspectly. As the National Journal wrote in 2007:
Top Democratic leaders and activists see Hispanic migration as a long-term opportunity for the party. The arrival of additional immigrant workers is “bad for blue-collars,” Rep. Barney Frank, D-Mass., chairman of the House Financial Services Committee, told National Journal late last year. But immigrants can help elect Democratic majorities, and “if [a Democratic Congress] were to significantly strengthen unions, then you would offset the negative effect on the income of workers,” he said.
There are two aspects if the transformative effect of mass immigration, First, as the quotes I just read boast, it serves to import more voters who will back statist, big-government politicians and policies, with the inevitably baleful consequences for economic growth and dynamism. But more deeply and permanently, mass immigration changes American society in ways that make statist solutions more attractive to voters at large, expanding the transformative effect beyond merely immigrant voters to the electorate at large.
First, votes. To exaggerate only a little, the left's immigration strategy is to replace the American electorate with one that is more receptive to statism. Overall, new immigrants are voting 60 to 80 percent in favor of the Democrats. It’s not just Hispanic immigrants either; there is evidence that Asian, African, European, and Middle Eastern immigrants are also voting mostly Democratic.
Over time, the political preferences of immigrant groups can change – assimilation translates into voting behavior more like Americans at large. For most immigrant groups, this means an increase in Republican voters. But for those few immigrant groups that, for historical reasons, have tended to vote heavily Republican, notably the Cubans, assimilation means that the Democratic share of their vote increases. In both cases, they come to be more like the rest of America. But this is a long, multigenerational process, and one that has been accelerated in the past by pauses in immigration, something that open borders would not allow.
In addition to the outsider status inherent to immigrants, which attracts them to the party of the pluribus rather than the unum, there are specific aspects of today’s immigration that make it easy for the statist to garner the lion’s share of the immigrant vote. For one thing, the large majority of immigrants are considered “minorities” under our current race laws and thus benefit from race-specific policies such as affirmative action, quotas, and minority set-asides. And the Left is the defender and promoter of such policies.
Furthermore, immigrants and their children disproportionately vote for the Left because they are disproportionately poor and, since those living in or near poverty generally pay no federal income taxes, the small-government message of lower taxes and slimmed-down government is simply not going to resonate with many immigrants voters.
A consequence of those lower incomes, as I've already mentioned, is greater reliance on government social programs. It just stands to reason that such voters will be drawn to the side that champions the expansion of welfare programs.
Of course, many immigrants don’t vote, either because they’re not citizens or they simply choose not to. But their children do vote, and among the children of immigrants we see trends – in addition to poverty and welfare use – that also make them more likely to be part of Eliseo Medina’s progressive coalition. Shockingly, 50 percent of births to native-born Hispanic women are illegitimate. Also, the research is clear that being married is one of the key indicators of voting Republican, but only 52 percent of native-born Hispanics ages 25 to 65 – largely the children and grandchildren of immigrants – are married, compared with 66 percent of native-born whites. Obviously, the breakdown of the family is a pressing problem for all Americans, but native-born Hispanics are among the hardest hit. Again, this makes it easier for the Left to get their votes.
Low incomes, welfare use, and family breakdown are only part of the problem. Almost all immigrants come from cultures that do not have a deep tradition of limited government, to put it mildly. This makes the Left’s approach of creating a patron-client relationship with the state more attractive – more natural – to many immigrants and their children, in contrast with the Right’s goal of a self-reliant citizenry holding government accountable. Survey research from the Pew Hispanic Center has shown that even those Hispanics who identify as Republicans are open to statism; about half of Latino registered voters who identified themselves as Republicans said they would rather pay higher taxes to support a larger government, compared to only 17 percent of white Republicans. In fact, Hispanic Republicans were more statist even than white Democrats.
But the electoral preferences of immigrant voters are only part of the story. Perhaps even more important is the fact that immigration creates a political, social, and economic environment more favorable to the Left. Political scientist Fredo Arias-King has written that “Immigration is a source of power for the étatist [statist] Left not so much because immigrants tend to vote for the party most associated with them, but because the consequences of immigration from poor countries fundamentally reinforces their argument for state intervention.” In other words, immigration transforms American society in ways that make native-born American voters more receptive to the arguments of the Left.
This is apparent in many policy areas. Take the issue of health insurance, for example. Nearly one-third of all uninsured people in the United States are immigrants or the young children of immigrants, and immigration is responsible for two-thirds of the growth in the number of uninsured in the past decade. Would the push for increased government involvement in health care have as much resonance with non-immigrant voters if immigration had not increased the seeming urgency of the problem?
Immigration also increases income inequality significantly, both by adding to the number of low-income people (immigrants and their young children account for one-fourth of all people in poverty in America) and by lowering wages for Americans on the bottom rungs of the labor market. As a result, calls for income redistribution and more social programs to address these problems are heard more sympathetically by native-born voters, even if they’re not poor themselves.
Finally, immigration increases support for big government by artificially increasing society’s diversity. A growing body of research, pioneered by Harvard’s Robert Putnam of Bowling Alone fame, shows that increased diversity results in a loss of civic engagement – not just between groups but even within groups. Immigration-driven increases in diversity mean everyone in the society is progressively less trusting and less involved in civic life, less likely to attend church, less likely to join the Masons, Hadassah, or the PTA, and less likely even to have friends over for dinner. And while this is undesirable in itself, what it means politically is that immigration contributes to the progressive weakening of what Edmund Burke called the “little platoons” of civil society, yielding more and more public space to government, and thus lending increasing credence to the arguments offered by the party of more government.
There's the old joke about the economist stranded on a desert island with only a can of beans and no way to remove the contents – so he assumes a can opener. Well, open borders belongs to the assume-a-can-opener parts of economics – it can be interesting and useful as a thought experiment, but it's place is in the sophomore year dormitory bull session, not in serious considerations of government policy.